Holes are too small. Fusion 360 Bore

Talk about all things CNC
Hans
Posts: 290
Joined: Tue Dec 25, 2012 5:19 am
Location: Connecticut, USA

Re: Holes are too small. Fusion 360 Bore

Post by Hans » Thu Sep 17, 2015 1:32 pm

There are two big differences from the pocket to the helical bore toolpath. One is the addition of a reaction force in +Z. If the holes are undersized in Y and correct in X, that would confirm that the problem is the spindle flexing by twisting the X rail because of this +Z force. The other difference is that by plunging away the center 3mm of the hole at each step, the pocket toolpath finishes with a 50% stepover whereas the bore toolpath is constantly engaged at 100%, which would cause the hole to be ~equally undersized in both axes.

Regardless, if you find good settings for full-depth finishing passes in your material, your problem will go away. A too-small finishing cut will not be effective in many materials because the tool just rubs instead of cutting when it comes in at a shallow angle. Conventional milling brings the cutting edge into contact from zero angle which produces less vibration for roughing cuts, but climb milling usually does a better job of starting the cut instead of just rubbing. Try a single finishing pass with a larger value (more like 10% stepover) in stock-to-leave. The results of changing these factors will vary depending on the workpiece material.
Shapeoko #1,123

Auarhau
Posts: 243
Joined: Tue Feb 25, 2014 8:46 pm

Re: Holes are too small. Fusion 360 Bore

Post by Auarhau » Thu Sep 17, 2015 2:51 pm

That's useful info Hans. I will incorporate it in my next test setup.

Fusion has two strategies that I can't tell the difference between.
The Circular strategy is used for milling cylindrical pockets and islands. The heights and depths are automatically derived from the selected cylindrical geometry which allows you to machine varying geometry in a single operation
The Bore operation allows you to bore mill cylindrical pockets and islands by selecting the cylindrical geometry directly. The heights and depths are automatically derived from the selected geometry which allows you to machine varying geometry in a single operation.
In addition it has normal pockets and 2d adaptive clearing. Not easy to know what to use..
ShapeOko 2. Nema 17 74 oz·in. GAUPS shield on Arduino Uno. DRV8825 Drivers x4 . Kress 1050 FME-1. Z Acme Screw. Threaded inserts table.

Fablicator
Posts: 113
Joined: Thu Jul 02, 2015 7:59 pm

Re: Holes are too small. Fusion 360 Bore

Post by Fablicator » Thu Sep 17, 2015 4:12 pm

Auarhau wrote:That's useful info Hans. I will incorporate it in my next test setup.

Fusion has two strategies that I can't tell the difference between.
The Circular strategy is used for milling cylindrical pockets and islands. The heights and depths are automatically derived from the selected cylindrical geometry which allows you to machine varying geometry in a single operation
The Bore operation allows you to bore mill cylindrical pockets and islands by selecting the cylindrical geometry directly. The heights and depths are automatically derived from the selected geometry which allows you to machine varying geometry in a single operation.
In addition it has normal pockets and 2d adaptive clearing. Not easy to know what to use..
EDIT: After experimenting some more circular uses distinct Z layers, while bore continuously lowers the Z path to form a helix.
Last edited by Fablicator on Thu Sep 17, 2015 5:38 pm, edited 1 time in total.

Auarhau
Posts: 243
Joined: Tue Feb 25, 2014 8:46 pm

Re: Holes are too small. Fusion 360 Bore

Post by Auarhau » Thu Sep 17, 2015 5:26 pm

Seems to me both do both of those.. http://fusion360.autodesk.com/learning/ ... 8ADF6B13B1

It was impossible to link directly to the correct references, so you have to click through fabrication to find the 2d cam resources.
ShapeOko 2. Nema 17 74 oz·in. GAUPS shield on Arduino Uno. DRV8825 Drivers x4 . Kress 1050 FME-1. Z Acme Screw. Threaded inserts table.

Fablicator
Posts: 113
Joined: Thu Jul 02, 2015 7:59 pm

Re: Holes are too small. Fusion 360 Bore

Post by Fablicator » Thu Sep 17, 2015 5:41 pm

Ahh, did some more experimentation and edited my previous post.

Bore uses a helical path, and circular uses distinct Z layers.

Bore:
Bore.jpg
Bore.jpg (63.18 KiB) Viewed 2589 times


Circular:
Circular.jpg
Circular.jpg (66.01 KiB) Viewed 2589 times

Hans
Posts: 290
Joined: Tue Dec 25, 2012 5:19 am
Location: Connecticut, USA

Re: Holes are too small. Fusion 360 Bore

Post by Hans » Thu Sep 17, 2015 6:37 pm

Right. The +Z force comes from the constant downward movement of the helical bore toolpath. That's what I was trying to say in my last post. The plunging and all the sharp corners in the circular pocketing toolpath make it a lot slower than the bore path, but it seems there is an accuracy trade-off from the extra force etc. A bore toolpath with a good finishing pass should combine the best of speed and accuracy.
Shapeoko #1,123

Auarhau
Posts: 243
Joined: Tue Feb 25, 2014 8:46 pm

Re: Holes are too small. Fusion 360 Bore

Post by Auarhau » Thu Sep 17, 2015 9:23 pm

A good discussion so far. I decided to try the ideas here, and many more in a little real life experiments. To try to get some number on how my machine does with the different strategies.
I made a test piece with 30, 6mm holes, using 2d pocket, bore and circular tool paths. All three strategies was tested with and without finish passes. With climb and conventional. And any combination. And I also tested boring conventional with 0.3mm stock to leave and then a contour/profile path at full depth to finish off the hole. I haven't gone through the numbers thoroughly yet. But so far a few things I've noticed.

It seems Gadgetman was on to something with single-cut operations vs choosing a roughing and finishing two part strategy. All my best holes (size, roundness and finish) were made by single operations that did the full diameter at once!

Secondly using conventional milling produced the best results. Some with climb milling was ok as well, but the general result was that anything concerning climb milling (be it complete single cut operations or used as finis passes) gave a little worse result.

I had little luck with the any variation of the circular path, and it also does a very annoying z up movement with every completed Z level.

The best two were simply 2D pockets (normal pockets) and bore cut as a single conventional milling operation. Those two produced the best looking holes and finish and was closest to spec. Of the two I think I prefer bore, since it's quicker to set up and quicker to cut.

The ones I did as a rough bore with a finishing contour pass came out pretty bad, too small and with lots of chatter.

It's a bit ironic that the one operation that did not produce good results, and caused me to start this thread, now seems the best option after I just removed the finishing passes.

To be honest none of my holes are perfect circles (but good enough for the girls I go out with!) and it's apparent that the Z flexes quite a bit in the x and y direction. Some of my holes were even slightly conical, and will bind at the bottom when I insert a pin (drill bit).

And yes, I am fully aware that measuring this with cheap calipers is not ideal. But I measured every hole many times to make sure I got repeatable results, and I checked the holes with a pin that is pretty damned close to 5,8mm to check the "real life fit".

I'll be back with more once I digest it all, and get some sleep! And I will also cut several more of the best ones, to confirm that they will do what I want repeatedly.
ShapeOko 2. Nema 17 74 oz·in. GAUPS shield on Arduino Uno. DRV8825 Drivers x4 . Kress 1050 FME-1. Z Acme Screw. Threaded inserts table.

AnonymousPerson
Posts: 758
Joined: Sun Apr 26, 2015 1:16 pm
Location: 3753 Cruithne

Re: Holes are too small. Fusion 360 Bore

Post by AnonymousPerson » Thu Sep 17, 2015 9:26 pm

Good stuff. :) Which material is that test in?

Kind of wondering if there will also be different best approaches for different materials. :)
Shapeoko 3 #516

Auarhau
Posts: 243
Joined: Tue Feb 25, 2014 8:46 pm

Re: Holes are too small. Fusion 360 Bore

Post by Auarhau » Thu Sep 17, 2015 9:48 pm

It was done in 12mm plywood. The kind with a hard brown surface on both sides. It's water resistant and was used for shelfs etc in my work wheicle. It seems a little tougher then normal ply. It was just what I had on hand. I'll post my files etc after I go through it all. Would be smart to test it on your own machine, because there are a lot of variables here. I think my Z axis is slightly weaker then some because of my special Z plate for my Kress. And this was with a 1/8 two flute straight bit, which is what I use most. I'm sure there will be differences in material. But wood is somewhere in the middle of the range I think, so should be safe for anything softer at least.
ShapeOko 2. Nema 17 74 oz·in. GAUPS shield on Arduino Uno. DRV8825 Drivers x4 . Kress 1050 FME-1. Z Acme Screw. Threaded inserts table.

WillAdams
Posts: 8542
Joined: Mon Apr 09, 2012 6:11 pm
Location: Pennsylvania --- south of the Turnpike, East of US-15
Contact:

Re: Holes are too small. Fusion 360 Bore

Post by WillAdams » Thu Sep 17, 2015 9:50 pm

Check to see that the V-wheels are tight?

Reduce the plunge rate?
Shapeoko 3XL #0006 w/ Carbide Compact Router w/0.125″ and ¼″ Carbide 3D precision collets
Nomad 883 Pro #596 (bamboo)

Post Reply